The objective of the proposed research is to evaluate ganglioside vaccines for the treatment of human melanoma. Two different antigen sources will be investigated: an anti-idiotype vaccine carrying the internal image of gangliosides and a glycoprotein vaccine having carbohydrate determinants common to gangliosides. During the last grant period we established a mouse monoclonal anti-idiotype antibody carrying the internal image of ganglioside GM3. The antibody has been humanized with a human constant region by genetic engineering technology. In the renewal application we will first evaluate the anti-tumor effect of the GM3 anti-id vaccine using an experimental melanoma model. A Phase I clinical study testing the immunogenicity of the humanized anti-id will be performed after in vivo tests with baboons as well as in vitro human systems demonstrate GM3 immunogenicity. We will develop human or humanized monoclonal anti-idiotype antibodies min-ticking other types of gangliosides as well. Development of other ganglioside vaccines is essential for the establishment of an effective immunotherapy targeted to ganglioside antigens since ganglioside expression varies considerably among different cancers as well as within the same histological type. We have previously developed human monoclonal antibodies to the melanoma-associated gangliosides, GM2 and GD2, which will be used as immunogens to develop anti-id monoclonal antibodies. Carbohydrate determinants of melanoma-associated gangliosides are known to be shared by glycoproteins of humans or animal species. The ability of the glycoprotein vaccine to induce anti-ganglioside immunity will be investigated. The source of glycoprotein vaccine will be either melanoma cells or tissues expressing the same epitope. Comparative studies among anti-id, glycoprotein, and ganglioside vaccine will determine the most effective form of vaccine in order to up-regulate ganglioside mediated anti-tumor immunity.